Design Methods for Complex Architectures (DEMCA)
Existing system development methods are insufficient to deal with the radically increased complexity stemming from increased scale, inter-dependencies, and types of ICT solutions. There is therefore an urgent need for a new system development paradigm. DEMCA will contribute to the development of the core elements of this paradigm: Design processes, architecture principles and governance approaches.
Contemporary Information and Communication Technology (ICT) solutions are increasingly inter-dependent and large scale i.e. complex. ICTs are no longer developed and managed as stand-alone and monolithic systems, but are built and implemented as parts of a larger whole. This change is particularly prominent in the domain of healthcare. As an example, the development and introduction of a new electronic patient record system at a hospital has to be aligned with existing work practices of doctors and nurses, with other existing clinical information systems (both electronic and paper based), and with patient administrative systems, laboratory information systems and imaging systems. The new system will be subject to both national and international standards for privacy regulation, message exchange and health indicator reporting. Thus, the new patient record becomes one small piece in a large architecture of information systems, medical devices, standards, governance structures, practices and users that together support patient handling and treatment. While systems developers may use existing systems development methods to develop the patient record component, aligning with and becoming a part of the larger “infrastructure” is a different challenge. This challenge is not about decomposing complicated requirements and deriving optimal technical designs, but grappling with the complexity that emerges from the heterogeneity of systems, actors involved, and the lack of control over other systems and components – in brief, the challenges are associated with a complex architecture.
This complexity poses a challenge to existing theoretical and methodological perspectives in informatics, since “the reductionism behind today’s software engineering methods breaks down in the face of systems complexity“ (Sommerville et al., 2012, p.71). Existing system development and software engineering methods, including state-of-the-art agile and user-centred methods as well as advanced programming tools and frameworks all seek to cope with the challenge of dealing with complexity. But most of them are software-centric and fail to address the heterogeneity of large scale, complex information systems. Such challenges are “unlikely to be addressed adequately by incremental research within established categories ... [but] ... require breakthrough research.” (Northrop et al., 2006, p. xi). We therefore need new conceptions of the nature of such systems and new approaches to design, develop, govern, monitor and assess them. In short, our vision is to develop a new system development paradigm. This paradigm must be founded on a theoretical understanding of complexity, to help design and govern large scale, and complex information systems.
The Global Infrastructures (GI) research group has over the last two decades worked towards developing a design-oriented “information infrastructure” theory, which has helped to shift the focus from standalone systems to large scale, complex information systems – or what we coin information infrastructures. Being centred on understanding and handling the complexity of information infrastructures, this theory has significant value academically, and practically for system developers and policy makers, and it has gained international acclaim within the Information Systems (IS) community with e.g. five research articles published since 2004 in MIS Quarterly. We believe this theory has the potential to be one core element in the required ‘paradigm change’ in system development methods. However, this theory needs further development, as well as further work on operationalizing it and producing more concrete system development methods. We define system development methods to include concepts, principles, taxonomies, models, frameworks, guidelines, and best practices, and in this document the term methods will refer to this broad range of outputs.
The GI research group is grounded in interpretively-informed, action-oriented research in the healthcare sector in developing countries, and multiple longitudinal studies in the healthcare sector in Norway. Our major empirical domain is healthcare, a complex and information-intensive sector undergoing major transformations that have significant impact on the use of ICT-based clinical, administrative and managerial systems. Thus, the healthcare sector is a highly relevant arena for research into the generic challenges emerging from the growing complexity of ICT solutions. For example, in Norway there are initiatives to redesign core aspects of the sector’s information infrastructure, e.g. to establish a national Care Summary Record to overcome the fragmentation of patient information being located in multiple disconnected systems. As another example, India has developed integrated systems to track every pregnancy and child birth in the country. Such initiatives encounter complexity in the form of multiple types of data, high volumes of information, technical heterogeneity of systems, a wide diversity of use and users, a politicized context where donors and governmental politics play a significant role, and a dynamic situation of rapid rate of change. Large scale, complex IS are implicated in a maze of legal, administrative and technical regulatory mechanisms, such as standards, professional norms and laws. Decision making within such a context needs to accommodate overlapping and perhaps conflicting interests of multiple stakeholders.
|DEMCA will leverage on the empirical research in the Global Infrastructures research group to advance methods targeted to the development of large scale, complex information systems.|
Within this empirical context (both in the Norwegian health sector and in developing countries) the DEMCA initiative and the involved PhD-students will conduct targeted research with the following objectives:
- Develop methods: The PhD-students will contribute to the development of system development methods for large scale, complex information systems. We will continue our work on the theme of design processes and build up a stronger focus on architecture and governance. The PhD students will work with these themes and have a particular focus on the operationalization of the new paradigm with methods to develop and manage large scale, complex information systems.
- Develop methods in the field: Within DEMCA, our objective is to conduct the required empirical research to test theories in the field, to produce methods that help us to practically deal with the complexity. We will do this with a basis in the opportunities offered by GI’s unique, real-world, and global research laboratory for conducting action-design research which has been on-going for two decades, centred on open-source-based health information infrastructures within the public health domain of various developing countries (called HISP). Similarly, we will draw on our established research in the Norwegian healthcare sector at department, hospital, regional and national level. Our objective is to leverage on the potential for mutual synergies between these two empirical domains in terms of developing new system development methods.
- Strengthen the linkages between research and teaching: The PhD students will bring methods that emerge from their action research into the teaching. This is new in the GI group, since we primarily have students with 3 year projects. We will evaluate, redesign and develop the current curriculum and courses offered by the GI group (bachelor, IFI-master, EVU-master and PhD), and contribute to make these methods part of other relevant courses and programmes at IFI. For example, students could already in the beginner courses (INF1000 and INF1050) be exposed to examples of large scale, complex information systems with related system development challenges and possible strategies and approaches.
- Strengthen the GI research group: DEMCA will contribute to other on-going activities aimed at strengthening the GI-group’s role as a central actor in the international network establishing this new paradigm (including research, teaching, innovation and conveying knowledge to the society at large) and to prepare us for the next round of SFF.